ABSTRACT BACKGROUND: One of the most feared complications with the use of cyanoacrylate for treatment of gastric varices is the occurrence of potentially life-threatening systemic embolism. Thus, endoscopists are turning towards new techniques, including endoscopic coiling, as a potentially safer and more effective treatment option. However, no studies have been performed comparing the two techniques. OBJECTIVE: This study aims to compare the safety and efficacy of endoscopic ultrasound guided coil and cyanoacrylate injection versus the conventional technique of injection of cyanoacrylate alone. DESIGN: A pilot randomized controlled trial. METHODS: Patients randomized into group I were treated with coil and cyanoacrylate, and those in group II with cyanoacrylate alone. Flow within the varix was evaluated immediately after the treatment session and one month following initial treatment. If thrombosis was confirmed, additional follow-up was performed 4 and 10 months following initial treatment. All patients underwent a thoracic computerized tomography scan after the procedure. RESULTS: A total of 32 patients, 16 in each group, were followed for an average of 9.9 months (range 1-26 months). Immediately after the procedure, 6 (37.5%) group-I patients and 8 (50%) group-II patients presented total flow reduction in the treated vessel (P=0.476). After 30 days, 11 (73.3%) group-I patients and 12 (75%) group-II patients were found to have varix thrombosis. In both groups, the majority of patients required only one single session for varix obliteration (73.3% in group I versus 80% in group II). Asymptomatic pulmonary embolism occurred in 4 (25%) group-I patients and 8 (50%) group-II patients (P=0.144). No significant difference between the groups was observed. CONCLUSION: There is no statistical difference between endoscopic ultrasound guided coils plus cyanoacrylate versus conventional cyanoacrylate technique in relation to the incidence of embolism. However, a greater tendency towards embolism was observed in the group treated using the conventional technique. Both techniques have similar efficacy in the obliteration of varices. Given the small sample size of our pilot data, our results are insufficient to prove the clinical benefit of the combined technique, and do not yet justify its use, especially in light of higher cost. Further studies with larger sample size are warranted.